perf(ocr): double batch sizes (detector=8, recognition=16)

4GB headroom in the container. Doubling batches should use ~2GB
more RAM but significantly speed up inference.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

docker-compose.yml

@@ -87,8 +87,8 @@ services:
       KRAKEN_MODEL_PATH: /app/models/german_kurrent.mlmodel
       OCR_CONFIDENCE_THRESHOLD: "0.3"
       OCR_CONFIDENCE_THRESHOLD_KURRENT: "0.5"
-      RECOGNITION_BATCH_SIZE: "8"
-      DETECTOR_BATCH_SIZE: "4"
+      RECOGNITION_BATCH_SIZE: "16"
+      DETECTOR_BATCH_SIZE: "8"
     networks:
       - archive-net
     healthcheck: